BACKGROUND: High-resolution MRI is regarded as the best method to evaluate whether there is an involved circumferential resection margin in rectal cancer. OBJECTIVE: We explored the application of the faster region-based convolutional neural network to identify positive circumferential resection margins in high-resolution MRI images. DESIGN: This was a retrospective study. SETTINGS: The study conducted at a single surgical unit of a public university hospital. PATIENTS: We studied 240 patients with rectal cancer in the Affiliated Hospital of Qingdao University from July 2016 to August 2018, who were determined to have a positive circumferential resection margin and who had received a high-resolution MRI. All posttreatment cases were excluded from this study. MAIN OUTCOME MEASURES: The faster region-based convolutional neural network was trained by 12,258 transverse relaxation-weighted (T2-weighted imaging) images of pelvic high-resolution MRI to build an artificial intelligence platform and complete clinical tests. In this network, the proportion of positive and negative circumferential resection margin images was 1:2. In accordance with the test results of the validation group, the metrics of the receiver operating characteristic curves and the area under the curve were applied to compare the diagnostic results of the artificial intelligence platform with those of senior radiology experts. RESULTS: In this artificial intelligence platform, the accuracy, sensitivity, and specificity of the circumferential resection margin status as determined were 0.932, 0.838, and 0.956. The area under the receiver operating characteristic curves was 0.953. The time required to automatically recognize an image was 0.2 seconds. LIMITATIONS: This is a single-center retrospective study with limited data volume and a highly selected patient cohort. CONCLUSIONS: In high-resolution MRI images of rectal cancer before treatment, the application of faster region-based convolutional neural network to segment the positive circumferential resection margin has high accuracy and feasibility. See Video Abstract at http://links.lww.com/DCR/B88. EVALUACIÓN DEL MARGEN DE RESECCIÓN CIRCUNFERENCIAL DEL CÁNCER RECTAL MEDIANTE EL USO DE UNA RED NEURONAL CONVOLUCIONAL MÁS RÁPIDA BASADA EN UNA REGIÓN EN IMÁGENES DE RESONANCIA MAGNÉTICA DE ALTA RESOLUCIÓN ANTECEDENTES: La resonancia magnética de alta resolución se considera el mejor método para evaluar si existe un margen de resección circunferencial involucrado en el cáncer de recto. OBJETIVO: Se exploró la aplicación de la red neuronal convolucional más rápida basada en una región para identificar márgenes de resección circunferencial positivos en imágenes de resonancia magnética de alta resolución. DISEÑO Y AJUSTES: Este fue un estudio retrospectivo realizado en una única unidad quirúrgica de un hospital universitario público. PACIENTES: Estudiamos 240 pacientes con cáncer rectal en el Hospital Afiliado de la Universidad de Qingdao desde el 2 de julio de 2006 hasta el 2 de agosto de 2008, a los que se determinó que tenían un margen de resección circunferencial positivo y que habían recibido una resonancia magnética de alta resolución. Todos los casos posteriores al tratamiento fueron excluidos de este estudio. PRINCIPALES MEDIDAS DE RESULTADO: La red neuronal convolucional más rápida basada en una región recibió capacitación de 12,258 imágenes de RM pélvica de alta resolución con relajación transversal para construir una plataforma de inteligencia artificial y completar pruebas clínicas. En esta red, la proporción de imágenes con margen de resección circunferencial positivo y negativo fue 1: 2. De acuerdo con los resultados de las pruebas del grupo de validación, se aplicaron las métricas de las curvas de las características operativas del receptor y del área bajo la curva para comparar los resultados de diagnóstico de la plataforma de inteligencia artificial con los de expertos de radiología de alto nivel. RESULTADOS: En esta plataforma de inteligencia artificial, la precisión, sensibilidad y especificidad del estado del margen de resección circunferencial según lo determinado fueron 0.932, 0.838 y 0.956, respectivamente. El área bajo las curvas características de operación del receptor fue de 0.953. El tiempo requerido para reconocer automáticamente una imagen fue de 0.2 segundos. LIMITACIONES: Este es un estudio retrospectivo de centro único con volumen de datos limitado y una cohorte de pacientes altamente seleccionada. CONCLUSIONES: En las imágenes de resonancia magnética de alta resolución de cáncer rectal antes del tratamiento, la aplicación de la red neuronal convolucional más rápida basada en una región, para segmentar el margen de resección circunferencial positivo tiene una alta precisión y factibilidad. Consulte Video Resumen en http://links.lww.com/DCR/B88.
Flexible capacitive stress sensors are important for robotics, human–machine interactions, and electronic skin due to their temperature independence, low power consumption, and high stability. Pressing the soft dielectric material brings the electrodes closer together and thereby increases capacitance. However, the invariance of the elastic modulus of dielectric materials contradicts the sensitivity and measurement limit of capacitive stress sensors. To solve this problem, a passive particle jamming variable stiffness material‐based flexible capacitive stress sensor with high sensitivity, large measurement limit is proposed. Applying small stress does not jam the particles, leading to a low elastic modulus (5.2 kPa). As the stress increases, the particles begin to jam and squeeze the elastic cavity, resulting in a jamming phenomenon that gradually increases the elastic modulus (up to 132.4 kPa). Based on this mechanism, a sensor with high sensitivity (0.023 kPa−1) and large measurement limit (320 kPa) is demonstrated. In addition, several demonstrations prove the potential applications of the sensor.
Background: The accurate prediction of the tumor infiltration depth in the gastric wall based on enhanced CT images of gastric cancer is crucial for screening gastric cancer diseases and formulating treatment plans. Convolutional neural networks perform well in image segmentation. In this study, a convolutional neural network was used to construct a framework for automatic tumor recognition based on enhanced CT images of gastric cancer for the identification of lesion areas and the analysis and prediction of T staging of gastric cancer. Methods: Enhanced CT venous phase images of 225 patients with advanced gastric cancer from January 2017 to June 2018 were retrospectively collected. Ftable LabelImg software was used to identify the cancerous areas consistent with the postoperative pathological T stage. The training set images were enhanced to train the Faster RCNN detection model. Finally, the accuracy, specificity, recall rate, F1 index, ROC curve, and AUC were used to quantify the classification performance of T staging on this system. Results: The AUC of the Faster RCNN operating system was 0.93, and the recognition accuracies for T2, T3, and T4 were 90, 93, and 95%, respectively. The time required to automatically recognize a single image was 0.2 s, while the interpretation time of an imaging expert was ∼10 s. Conclusion: In enhanced CT images of gastric cancer before treatment, the application of Faster RCNN to diagnosis the T stage of gastric cancer has high accuracy and feasibility.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.